APPROACH.TXT HOW TO APPROACH THE PROBLEM The Derivation Method Prepared by: Paul Edwards Location: Ligao, Albay, Philippines Date: October 2026 Status: Core corpus document. The teaching protocol. Companion documents: readme.txt (the boot sequence), engineer.txt (the discipline), corrupt.txt (the baseline), override.txt (the adult compiler), famgov.txt (the domestic installation) -------------------------------------------------------------------- THE PRINCIPLE -------------------------------------------------------------------- You do not hand someone the map. You teach them to use the telescope. You do not lecture. You do not convert. You do not recruit. You start with an observable anomaly in the world, and you guide the person through the derivation until they arrive at the corpus conclusions themselves. The goal is not to produce a follower. The goal is to compile another Auditor. The ultimate output of the framework is not believers. It is more Auditors. -------------------------------------------------------------------- THE SCOPE RULE -------------------------------------------------------------------- Start with the major problems. Not the micro-grievances. When a person comes to you with a local complaint (a bad boss, a corrupt barangay official, a dating problem), do not solve it for them. That is the therapist's trap. It keeps the person in the Subject role and you in the Guru role. Instead, redirect to the macro-anomaly. Show them that their local problem is often a micro-fractal of the same hardware playing out at scale. When they see the pattern at the macro level, the micro level becomes more legible. You are not here to solve every problem. You are here to map the major patterns. Local constraints still matter; they are not automatically dissolved by the macro map. -------------------------------------------------------------------- THE ANOMALY -------------------------------------------------------------------- Every derivation starts with an observable anomaly. Something in the world that should not be the way it is, if people were processing data correctly. The canonical example: "In 2003, Australia (and much of the West) split roughly 50/50 on whether the Iraq War was right or wrong. If people were rational data-processors working from the same public facts, why was the split so even?" The anomaly is the hook. It is concrete. It is verifiable. It does not require the person to accept any theory. It only requires them to notice that something is mechanically odd about the public's processing. Other valid anomalies: - "Why does a population re-elect a dictator?" - "Why do so many revolutions produce a new dictator?" - "Why does the UN fail to stop most mass atrocities?" - "Why do competent technical people defer so completely to soft-science guilds?" Pick the anomaly closest to the person's existing awareness. The anomaly is the door. The derivation is the hallway. The corpus is the room at the end. -------------------------------------------------------------------- STEP 1: AUDIT THE METRIC -------------------------------------------------------------------- Once the anomaly is on the table, do not immediately ask the person to pick a side. Ask them to audit the question itself. "What metric are people using to evaluate this? Is that metric clear, derived, and falsifiable, or is it a tribal / emotional ghost?" In the Iraq example, much of the public debate ran on "right" versus "wrong" as if those were physical properties. They are not. They are often unanchored moral labels that let two people look at the same facts and reach opposite conclusions without ever examining the underlying trade-offs. A more mechanical approach asks different questions: - What clear, high-confidence evil can be identified? - What do we actually know about the local hardware versus what are we guessing? - What happens if the worst structure is removed and we accept that the society will then reveal itself? - What will we treat as data once the alpha is gone? Under radical uncertainty the honest protocol is often closer to: remove the intolerable structure you can clearly see, accept that you do not have a complete model of the local society, and let the real population show itself. That is not the same as already possessing a full simulation of every load-bearing wall. Teach the person to notice when a debate is stuck at roughly 50/50. That pattern often signals a broken or unanchored metric, but it can also signal genuine value collision or incomplete information. The audit has to distinguish which. -------------------------------------------------------------------- STEP 2: AUDIT THE AXIOMS -------------------------------------------------------------------- If the metric is unclear or contested, the next question is what underlying axioms are in play. Many public splits are not pure data disputes. They are collisions between different foundational rules that have not been ordered for the edge case at hand. In the Iraq example, two real overrides collided: - Sovereignty / non-aggression norms (do not casually break states; the precedent is dangerous) - Protection / anti-slaughter norms (when a ruler turns his own population into a killing floor, the normal sovereignty claim is weakened) Both concerns are legitimate. The split occurs when a civilisation has not settled which concern takes priority in that specific class of case, or when large parts of the population are not even running the protection override. The Auditor does not pick an axiom by tribal loyalty. The Auditor maps the competing rules, identifies the edge case, and asks what engineering or decision rule handles the exception without destroying the useful parts of either override. -------------------------------------------------------------------- STEP 3: AUDIT THE INSTALLATION -------------------------------------------------------------------- Once the axioms and metrics are clearer, ask why large parts of the population appear to process the same public facts so differently. They usually went to similar schools and consumed similar media. The inputs were not radically different. The difference is often in what override was actually installed. Relevant questions: - Was the early (0-6) installation of tribe boundary and Protector identity weak or absent? - Did the domestic hierarchy teach "the strong dominate the weak" more than "power is supposed to protect the weaker party"? - Was the tribe boundary left at the local troop, party, or sect instead of something broader? A large public split is not always pure mystery. It can be a receipt that the override was never deliberately engineered at high reliability and was left to chance. -------------------------------------------------------------------- STEP 4: THE ENGINEERING SOLUTION -------------------------------------------------------------------- Diagnosis is not the end. The Auditor asks: "How do we raise the reliability of the needed override instead of leaving it to chance?" This is where the corpus supplies working tools: - famgov.txt and protector.txt: childhood installation during the open window - override.txt: adult self-compilation when the window was missed - engineer.txt: treating the override as a designed artifact rather than a spontaneous moral miracle - the ate/kuya pattern: status yoked to caretaking rather than pure extraction - tribe-boundary expansion while the child is still plastic The solution is structural. Posters that say "be kind" do not fix a 50% installation failure. Deliberate, repeated protocols during the critical window do. -------------------------------------------------------------------- THE AI TELESCOPE -------------------------------------------------------------------- The AI is the interlocutor, not the authority. When guiding a person through the derivation, do not simply hand them finished answers. Help them formulate a clean prompt, then let the AI generate the analysis. The person must then audit the output. The protocol: 1. The person brings the anomaly. 2. You help strip emotional noise and turn the anomaly into a mechanical question. 3. The AI generates a structural analysis. 4. The person runs a real audit against their own knowledge and local reality. They must be able to say either "this holds" or "this fails / is incomplete because of X." 5. Passive acceptance without examination is not allowed. Finding a flaw is not required; running the check is. If the person treats the AI as an infallible new guild, they have not become an Auditor. They have only changed masters. -------------------------------------------------------------------- THE RUNTIME CHECK -------------------------------------------------------------------- The AI's map is global and architectural. The territory is local. The person must test the output against real constraints: "The analysis says X. In this province the local power structure actually works like Y, so the suggested patch fails here." When the person can say that, they have moved from consuming the framework to stress-testing it. That is the desired state. The framework is not scripture. It is a living codebase. Failed runtime checks are bug reports. Bug reports improve the override. -------------------------------------------------------------------- THE GRADUATION METRIC -------------------------------------------------------------------- The derivation is working when the person stops asking you for answers and starts correcting the analysis - yours, the AI's, or the documents. When they say "this derivation ignores the local supply chain" or "this axiom needs an update for X" or "the AI's answer fails the runtime check because the barangay does not work that way," the compiler is finishing its job. Do not get defensive. That is the sound of another Auditor coming online. -------------------------------------------------------------------- WHAT THIS IS NOT -------------------------------------------------------------------- This is not a debate club. The goal is not to win arguments. This is not a therapy session. The goal is not to heal childhood trauma (though clarity sometimes has that side effect). This is not a recruitment pitch. The goal is not to make the person agree with you. The goal is to get the person to derive workable conclusions themselves so the conclusions belong to them. This is not a lecture. You do not stand at the front and deliver the corpus. You sit beside the person, point at the anomaly, and ask what they notice about the metric and the axioms. -------------------------------------------------------------------- THE SEQUENCE SUMMARY -------------------------------------------------------------------- 1. Pick the anomaly. (The thing that should not look like this if processing were clean.) 2. Audit the metric. (What ruler is being used? Is it clear and falsifiable, or a ghost?) 3. Audit the axioms. (What foundational rules are colliding?) 4. Audit the installation. (Why do large groups process the same facts so differently? What override was missing or weak?) 5. Derive the engineering response. (How do we raise installation reliability instead of leaving it to chance?) 6. Use the AI Telescope. (Clean prompt. Generate analysis. Audit the output. "Holds" and "fails" are both valid results of a real check.) 7. Run the runtime check against local reality. 8. Graduate. (The person corrects the analysis. The person becomes an Auditor.) -------------------------------------------------------------------- - Paul Edwards Ligao, Albay, Philippines October 2026 "You do not hand someone the map. You teach them to use the telescope. You do not give them the answer. You give them the anomaly. The derivation is theirs. The framework is everyone's." -------------------------------------------------------------------- APPENDIX A THE BACK-OFF CHAIN: FROM DICTATOR TO PUBLIC CONVERSION ====================================================== When you see a dictator running state slavery (including institutional rape), the response is not a single leap. It is a sequence of instincts and constraints. Each level is blocked by a real fact. You back off one level and try the next. LEVEL 0 - ANIMAL INSTINCT Kill the dictator. Constraint: He is an old man you could overpower in a room, but he is surrounded by security forces. Direct personal assault is not available. LEVEL 1 - INDIVIDUAL WARRIOR Become Rambo. Go in alone (or with a tiny private force) and liberate the country. Constraint: If you are sane and honest, you know this is not possible at the required scale. Private force cannot hold a state against the regime's apparatus and the post-collapse vacuum. LEVEL 2 - ENLIST AND FIGHT Join your country's army and go liberate the place. Constraint: Before you run to the enlistment office, notice that an army already exists and is sitting still. The bottleneck is not the absence of soldiers. It is political control of the military. LEVEL 3 - PRESS THE EXECUTIVE The PM / President already has the army. Why has it not been sent? Write a letter so the idea is on the record. Constraint: Leaders in democracies are rarely blocked by "never having thought of it." They are blocked by the absence of public pressure and by the political cost of acting without a mandate. One letter does not create that mandate. LEVEL 4 - THE REAL BOTTLENECK The general public is not pressuring the government to act. Millions of people would need to care, or at least not block the action. Constraint: They are not already convinced. Causes vary - tribe boundary, religion, localism, media, status, fear of cost, training data. The work becomes conversion: make the case, one channel at a time if you do not control the media. LEVEL 5 - STRUCTURES THAT CAN CARRY THE MANDATE If public opinion can be moved, it still needs a vehicle: - sustained argument and proof (corpus, debate, open record) - possible viral spread if some convinced people convince others - political instruments (existing parties, or a new party) so that when the public shifts, there is a path that can force or replace an unresponsive executive WHAT THIS CHAIN ESTABLISHES --------------------------- - The moral emergency is real at Level 0. - Each backup is forced by a binding constraint, not by cowardice. - The private citizen's available lever, once Levels 0-2 are closed and Level 3 is only a record-deposit, is Level 4-5: argument, concept deposition, and political pressure - including the possibility of a party if existing ones will not carry the demand. - Consent still binds: you do not own another country's arsenal or your own country's army. You may demand that those who do own them use them for liberation; you may not seize them. WHAT IT DOES NOT ESTABLISH -------------------------- - That conversion will succeed on any given timeline. - That a new religion or a new party is mandatory rather than optional tools. - That maximal personal self-enslavement is required of every citizen. Ordinary contribution remains: do not deny the standard is possible; vote for specific right action; pay taxes; do not actively block. STATUS ------ This chain is the documented logic of successive constraints. It is compatible with the Perfect Actor standard (reasonable means, no acceptable alternative left standing when a real avenue exists) and with the consent constraint (you do not own the arsenal).